Co-formulated elvitegravir, cobicistat, emtricitabine, and tenofovir disoproxil fumarate versus ritonavir-boosted atazanavir plus co-formulated emtricitabine and tenofovir disoproxil fumarate for initial treatment of HIV-1 infection: a randomised, double-blind, phase 3, non-inferiority trial.

BACKGROUND The HIV integrase strand transfer inhibitor elvitegravir (EVG) has been co-formulated with the CYP3A4 inhibitor cobicistat (COBI), emtricitabine (FTC), and tenofovir disoproxil fumarate (TDF) into a once-daily, single tablet. We compared EVG/COBI/FTC/TDF with a ritonavir-boosted (RTV) protease inhibitor regimen of atazanavir (ATV)/RTV+FTC/TDF as initial therapy for HIV-1 infection. METHODS This phase 3, non-inferiority study enrolled treatment-naive patients with an HIV-1 RNA concentration of 5000 copies per mL or more and susceptibility to atazanavir, emtricitabine, and tenofovir. Patients were randomly assigned (1:1) to receive EVG/COBI/FTC/TDF or ATV/RTV+FTC/TDF plus matching placebos, administered once daily. Randomisation was by a computer-generated random sequence, accessed via an interactive telephone and web response system. Patients, and investigators and study staff who gave treatments, assessed outcomes, or analysed data were masked to the assignment. The primary endpoint was HIV RNA concentration of 50 copies per mL or less after 48 weeks (according to the US FDA snapshot algorithm), with a 12% non-inferiority margin. This trial is registered with ClinicalTrials.gov, number NCT01106586. FINDINGS 1017 patients were screened, 715 were enrolled, and 708 were treated (353 with EVG/COBI/FTC/TDF and 355 with ATV/RTV+FTC/TDF). EVG/COBI/FTC/TDF was non-inferior to ATV/RTV+FTC/TDF for the primary outcome (316 patients [89·5%] vs 308 patients [86·8%], adjusted difference 3·0%, 95% CI -1·9% to 7·8%). Both regimens had favourable safety and tolerability; 13 (3·7%) versus 18 (5·1%) patients discontinued treatment because of adverse events. Fewer patients receiving EVG/COBI/FTC/TDF had abnormal results in liver function tests than did those receiving ATV/RTV+FTC/TDF and had smaller median increases in fasting triglyceride concentration (90 μmol/L vs 260 μmol/L, p=0·006). Small median increases in serum creatinine concentration with accompanying decreases in estimated glomerular filtration rate occurred in both study groups by week 2; they generally stabilised by week 8 and did not change up to week 48 (median change 11 μmol/L vs 7 μmol/L). INTERPRETATION If regulatory approval is given, EVG/COBI/FTC/TDF would be the first integrase-inhibitor-based regimen given once daily and the only one formulated as a single tablet for initial HIV treatment. FUNDING Gilead Sciences.